Abstract
Background/Aim: Stage migration, a phenomenon triggered by technological advancements allowing more sensitive tumor spread detection, results in alterations in the distribution of cancer stages within a population. Canine multicentric lymphoma is staged I to V based on the affected anatomic site(s) and substage a or b depending on the presence of tumor-related clinical signs. The primary objective of this study was to assess the influence of various diagnostic techniques on staging accuracy and determine whether multiple staging methods lead to significant stage migration, impacting the reliability of disease stage assignments. Materials and Methods: Dogs cytologically diagnosed with multicentric lymphoma were staged using four different staging methods (A-D): A (physical examination, hemogram, blood smear), B (A plus thoracic X-ray, abdominal ultrasound), C (B plus liver and spleen cytology) and D (C plus bone marrow cytology). Results: Twenty-three dogs were enrolled: 16 females (70%) and seven males (30%). Regarding immunophenotype, 21 dogs (91.3%) were B-cell and two dogs (8.7%) were T-cell. Stage migration was observed between all staging methods. Between A and B, 12 animals migrated from stage III to stage IV. Between B and C, four animals migrated, three to a higher stage (stage III to IV) and one to a lower stage (stage IV to III). Between C and D, one animal migrated from stage IV to V. The differences between staging methods A and B were statistically significant (p≤0.001). Conclusion: Stage migration in canine multicentric lymphoma depends on the diagnostic methods used and reinforces the need to use standardized staging methods to avoid it.
Lymphoma is the most common hematopoietic neoplasm in dogs, representing approximately 83% of all canine hematopoietic malignancies (1, 2). This fact combined with its high chemo-sensitivity, with complete remission rates ranging from 65 to 90%, makes it one of the most commonly treated cancers in dogs (3).
The majority of canine lymphomas are diffuse and intermediate to large cell type, which makes it possible, in most cases, to obtain a precise diagnosis through cytology. Lymphoma’s staging follows the classification system outlined by the World Health Organization. This system categorizes dogs from stage I to V depending on the affected anatomic site(s) and further refines the classification into substage a or b based on clinical presentation (4).
Staging dogs with cancer is crucial for making informed decisions regarding therapy, tracking treatment response, determining prognosis, and categorizing dogs for participation in clinical trials. Accurate staging is essential to evaluate and interpret results from clinical studies, to assess the effectiveness of different treatment protocols and to establish the prognostic significance of various parameters, including stage. To date there is no consensus stating the diagnostic tests needed to be performed to stage a patient which can result in stage migration. In truth, the use of newer, more advanced, and sensitive staging tests enables the detection of diseases that might have gone unnoticed with simpler diagnostic techniques (5).
Stage migration refers to the shift to either a lower or higher disease stage resulting from the utilization of more sensitive staging tests, such as ultrasound, radiographs, PET scans, and bone marrow cytology (5). Initially described by Feinstein et al. (6) in 1985 for human lung cancer, this phenomenon has been documented across various human tumors, including prostate tumors (7), breast cancer (8), Hodgkin’s lymphoma (9), and laryngeal cancer (10). In veterinary oncology, the same phenomenon has been observed and reported in several tumors, including lymphoma (5, 11-13). However, in addition to being relatively few, previous published studies incorporate heterogenous patient populations. Furthermore, there is a lack of standardization in the complementary exams performed on each animal, potentially introducing bias to the results (5).
The objective of this study was to assess the influence of various diagnostic techniques on the staging of canine multicentric lymphoma within a uniform population subjected to a standardized set of diagnostic procedures. The study aimed to investigate whether employing multiple staging methods in dogs with lymphoma could lead to significant stage migration.
Materials and Methods
Case selection. The animals included in the study were dogs referred to the veterinary hospital of Auna Especialidades Veterinárias, between September 2016 and June 2017.
Inclusion criteria included: 1) >1 year of age; 2) newly diagnosed multicentric diffuse large B-cell lymphoma (DLBCL, centroblastic or immunoblastic) or T-cell lymphoma; 3) all animals had to be naïve to treatment for the current cancer; 4) owners’ cooperation regarding the performance of all complementary exams required.
Diagnosis workup. All dogs suspected of having multicentric lymphoma, evidenced by peripheral lymphadenopathy, underwent a complete physical examination including palpation and measurements of the longest diameter of all assessable lymph nodes required by the RECIST lymphoma criteria (14). On physical examination, dogs with enlarged organs (e.g., liver, spleen) were considered positive for lymphoma in these organs. The same set of complementary exams was performed in every animal and this included: fine needle aspiration (FNA) of one or two affected lymph nodes for flow cytometry analysis and cytological evaluation; thoracic x-ray (three views: right lateral, left lateral and ventrodorsal); abdominal ultrasound; ultrasound-guided FNA of the liver and spleen; bone marrow aspirate and cytology obtain from costochondral junctions of the ribs; complete and differential blood cell count; serum biochemical profile including, at least, total proteins (TP), albumin (ALB), creatinine (CREA), urea nitrogen (BUN), alanine aminotransferase (ALT), alkaline phosphatase (AP), and calcium (Ca); blood smear.
The lymph node samples obtained for flow cytometry and immunophenotyping were transferred to a tube with NaCl solution mixed with a drop of autologous serum, obtained by blood centrifugation, and stored in refrigeration before being sent to Barcelona Autonoma University, where the tests were performed. Blood smear was considered positive when abnormal circulating lymphoblast’s where observed in the sample (5, 6).
Abdominal organs were considered affected in the ultrasound examination if they revealed echogenicity changes and/or enlargement, previously reported as typical of abdominal lymphoma (15-17). Cytological samples were obtained from the liver and spleen of all the animals, regardless of their ultrasonographic appearance.
All cytologic samples from the liver and spleen were evaluated for sample quality and presence of immature lymphoid cells. The smears were immediately stained and evaluated after sample collection. Samples were excluded if more than 50% of the nucleated cells were lysed and, in liver samples, if no hepatocytes were present (15). In these situations, a new sample was collected. Cytological diagnosis of lymphoma infiltration in the liver was based on the finding of many large lymphocytes in samples that were moderate to highly cellular and contained hepatocytes. In the spleen, cytological diagnosis of lymphoma was based on the finding of a marked predominance of large lymphocytes (more than 50% of all lymphocytes seen) (15). The bone marrow was considered affected if more than 20% lymphoblasts were observed on cytological evaluation (18).
Clinical staging. Each dog was staged based on the results of physical examination and diagnostic workup according with the World Health Organization’s clinical staging system for lymphoma in domestic animals (19). The stage was determined using four distinct stage grouping methods (A to D) (Table I). Results of imaging studies (abdominal ultrasound and thoracic x-rays) were considered dominant over physical examinations findings for the determination of stage. Likewise, cytological findings were considered more relevant for staging than imaging findings (5, 16).
Staging methods.
Statistical analysis. Statistical analysis was performed using the SPSS software version 24 (IBM, Armonk, NY, USA). McNemar-Bowker test was used to assess the difference between the staging method groups. Values of p<0.05 were considered significant.
Results obtained from ultrasonographic and cytological evaluation were crossed to evaluate sensitivity, specificity, positive and negative predictive values and accuracy (20, 21) for the detection of hepatic and splenic lymphoma infiltration in each organ.
True positives were defined as organs with both ultrasonographic and cytology results compatible with lymphoma. False positives were classified as organs with abnormal ultrasonographic appearance without a diagnosis of lymphoma in cytology. True negatives were defined as organs with both normal ultrasonographic and cytologic appearance. False negatives were classified as organs with normal ultrasonographic appearance but with a diagnosis of lymphoma on cytology.
Results
Patient characteristics. Twenty-three dogs were enrolled in the study: 16 females (70%) and seven males (30%). Regarding the neuter status, 14 dogs (60.9%) were castrated (10 females and four males) and nine dogs (39.1%) were intact (three males and six females). Breeds included were mixed breed (three dogs; 13%), Schnauzer (one dog; 4.3%), crossed German Shepherd (two dogs; 8.7%), Beagle (one dog; 4.3%), French Bulldog (five dogs; 21.7%), Yorkshire (two dogs; 8.7%), Pitbull (one dog; 4.3%), German Shepherd (one dog; 4.3%), Maltese (one dog; 4.3%), Labrador Retriever (one dog; 4.3%), Golden Retriever (one dog; 4.3%), Pincher (one dog; 4.3%), Malinois (one dog; 4.3%), Belgian Shepherd Tervuren (one dog; 4.3%) and crossed Golden Retriever with Belgian Shepherd (one dog; 4.3%). Patients age ranged from 4 to 13 years old, with a mean age of 8.6±2.4 years. Regarding the body weight, values ranged between 3 kg and 40.8 kg.
Regarding immunophenotype, 21 dogs (91.3%) were B-cell and two dogs (8.7%) were T-cell. Morphologic classification according with Kiel classification was available in 18 animals, being 13 immunoblastic (72%) and five centroblastic (28%).
Staging tests. On physical examination, 23 dogs (100%) had peripheral lymph node enlargement on both sides of the diaphragm. One dog (4.3%) presented pronounced facial edema due to submandibular lymph node enlargement. Two dogs (8.7%) revealed edema of the posterior limbs and scrotum and one dog (4.3%) presented bladder distention due to inguinal lymph node enlargement. Hepatomegaly and splenomegaly were detected in two (8.7%) and five (21.7%) dogs, respectively.
Complete blood count showed abnormalities in 17 dogs (73.9%). The two most common cytopenias observed were anemia (six dogs; 26.1%), and thrombocytopenia (six dogs; 26.1%). All low platelet count results were confirmed with a blood smear to assure that platelet clumps were not present. Three dogs (13%) presented lymphocytosis and 10 dogs (43.5%) revealed monocytosis. Leukocytosis with neutrophilia was detected in three dogs (13%) and lymphopenia was present in four dogs (17.4%). Only one dog (4.3%) presented abnormal circulating lymphoblasts after blood smear evaluation.
Blood biochemistry revealed abnormalities in 14 dogs (60.9%). Increased liver enzymes were the most common abnormalities found, with seven dogs (30.4%) presenting increased ALT and nine dogs (39.1%) increased AP. Other abnormalities found were hypercalcemia (three dogs; 13%), increased creatinine (two dogs; 8.7%), increased BUN (one dog; 4.3%), hypoproteinemia with hypoalbuminemia (one dog; 4.3%), hypoglycemia (one dog; 4.3%) and hyperglycemia (two dogs; 8.7%).
On bone marrow aspirate evaluation, two dogs (8.7%) had bone marrow infiltration.
Thoracic X-rays revealed lymphadenopathy in six (26.1%) animals. In addition to lymphadenopathy, one dog (4.3%) also presented pleural effusion. No significant statistical association was found between thoracic X-ray abnormalities and the development of respiratory clinical signs (p=0.127).
Abdominal ultrasonography revealed abnormalities in 18 dogs (78.2%). Among these, 17 animals (94.4%) showed abnormal appearance of the liver and/or spleen and two animals (11.1%) abdominal lymph nodes enlargement. Ultrasonographic and cytologic results for the liver and spleen are summarized in Table II. Ultrasonographic appearance of the liver was abnormal, with diffuse heterogeneous appearance of the parenchyma, in five of the 23 dogs (21.7%). Comparing ultrasonographic and cytological results, it was possible to observe that all five positive cases in the ultrasonographic exam were confirmed as positive by cytological examination (true positives). In opposition, 10 of the 18 negative cases in the ultrasonographic examination, were later proved as positive in cytology, being these 10 false negative cases. For abdominal ultrasound examination alone, the following values were obtained: sensitivity 33.3%, specificity 100%, positive predictive value100%, negative predictive value 44.4% and accuracy 56.5%.
Detection of liver and spleen involvement by ultrasonography vs cytology (ultrasound guided).
Abnormal ultrasonographic appearance of the spleen, with hypoechoic areas, moth eaten or coarse parenchyma patterns, were observed in 17 of the 23 animals (73.9%). Nevertheless, only 15 of these cases were confirmed, by cytological evaluation, as true positives, and two cases were classified as false positives. Moreover, three dogs classified as negative in the ultrasonographic examination were later discovered as being positive by cytological examination. The ultrasonography of the spleen showed sensitivity of 83.3%, specificity of 60%, positive predictive value of 88%, negative predictive value of 50% and accuracy of 78.3%.
Assessment of stage according to each staging method. The staging results obtained with each staging method are illustrated in Figure 1. Stage migration (reclassification of stage) was observed between all staging methods (Figure 2). Using the method B, 12 dogs were re-staged to a higher stage (III to IV), due to abnormal appearance of the liver and/or spleen on ultrasonography examination in the absence of organomegaly detected on physical examination. One dog was downstaged (IV to III) in method C because of a negative spleen cytology in a dog whose spleen ultrasound appearance was considered positive. Moreover, three dogs were transferred to a higher stage (III to IV) due to absence of abnormal ultrasonography appearance of the spleen and liver with a positive cytological result. In method D, one dog was re-staged (IV to V) due to positive bone marrow cytology. The difference between staging methods A and B was statistically significant (p=0.001). No significant statistical differences were found between staging methods B and C and staging methods C and D (p>0.05) (Figure 2).
Percentage of dogs in stages III, IV, and V when staged using A (physical examination, hemogram, blood smear), B (A plus thoracic X-ray, abdominal ultrasound), C (B plus liver and spleen cytology), and D (C plus bone marrow cytology) staging methods.
Stage migration between staging methods (n=23): A (physical examination, hemogram, blood smear), B (A plus thoracic X-ray, abdominal ultrasound), C (B plus liver and spleen cytology), D (C plus bone marrow cytology).
Discussion
Clinical staging in canine lymphoma is a crucial process that provides vital insights into the disease’s extent and characteristics, serving as a guide for treatment decisions and offering valuable prognostic information to both veterinarians and pet owners. Once a conclusive lymphoma diagnosis is established, staging becomes paramount in assessing the distribution and degree of the oncologic disease. The majority of published studies include populations of animals that underwent different diagnostic and staging workup, as well as animals with different lymphoma subtype (22-25). Therefore, the primary goal of this study was to determine whether different staging methods in dogs with multicentric lymphoma result in significant stage migration in a homogeneous population submitted to the same set of diagnostic procedures. Of the 23 dogs enrolled in the study, during physical examination, hepatomegaly was only detected in two animals and splenomegaly in five. However, the introduction of more advanced diagnosis tests revealed a larger number of animals (18 dogs) with the involvement of intraabdominal organs, which enhances the fact that physical examination, although important, is a highly insensitive and subjective method for lymphoma staging. The same conclusion was reached by Flory et al. (5) in their study with 59 dogs, where only 30% of the cases had liver and/or splenomegaly detectable on physical examination, while ultrasonographic changes of the liver and spleen were encountered in 43% and 74% of the animals, respectively.
Diagnosis imaging techniques have allowed a substantial improvement in the assessment of the extension of the disease in the body. Thoracic radiographs are important to determine involvement of intrathoracic structures. In this study, 26.1% of the dogs presented detectable abnormalities in the thoracic x-rays. The incidence of thoracic radiograph abnormalities observed was lower than that reported in previous studies by Starrak et al and Blackwood (27), where the encountered incidence was 71% and 76% respectively. No stage migration was observed with the inclusion of thoracic X-rays to the group of diagnosis tests.
Abdominal ultrasonography is another useful and commonly performed imaging technique, which can aid in the determination of the involvement of intraabdominal organs. In this study the positive predictive value for hepatic ultrasonography was high (100%), which seems to indicate that ultrasonography is a good test for detection of liver involvement in dogs with lymphoma. However, the negative predictive value was low (44.4%), indicating that a negative result on liver ultrasonographic examination should always be confirmed with a cytological examination. These results contrast with the results previously reported in another study by Crabtree et al. (15), where both the positive and negative predictive values were high enough to consider ultrasonography a useful test for the detection of hepatic involvement.
Regarding the spleen, similarly to what was observed in the liver, the ultrasonography positive predictive value was high (88%), while the negative predictive value was low (50%). Again, these results contrast with the ones previously documented by Crabtree et al. (15), where the negative predictive value for splenic ultrasonography was so high (100%) that the authors considered unnecessary the performance of FNA for cytological evaluation of an ultrasonographic normal spleen. In opposition, in the same study, the positive predictive value and the specificity were low (64.67% and 23.3% respectively), which led to the conclusion that cytology should be always performed when an abnormal ultrasonographic image of the spleen is detected to assess whether or not lymphoma is present (15). In our work, the addition of abdominal ultrasonography resulted in a significant stage migration, with 12 dogs shifting from stage III to stage IV. Similar results have been reported by Flory et al. (5) where the addition of abdominal sonography was accompanied by a shift into higher stages.
FNA is a safe and effective method for obtaining cell samples for microscopic examination, which makes it a good-to-excellent technique for obtaining high-quality diagnostic specimens (28). With the association of cytological examination to the ultrasound, three animals changed to a higher stage (III to IV) and one dog changed to a lower stage (IV to III). This downward stage migration has been reported in previous articles (11, 15) where the addition of cytology to abdominal ultrasound resulted in the confirmation of negative cases previously diagnosed as false positive during imaging evaluation. Nevertheless, in accordance with other studies, the inclusion of cytology resulted in a higher number of shifts into higher stages (11, 15).
Bone marrow evaluation is important to assess lymphoma stage, however, there are no standard techniques for quantifying bone marrow infiltration. The inclusion of this step in the staging procedure resulted in the migration of one dog to a higher stage (IV-V). Some researchers defend that bone marrow evaluation is only required when peripheral blood abnormalities, such as cytopenias, suggest a strong infiltration (14). In this study both dogs classified as stage V revealed peripheral blood abnormalities, with one animal presenting grade 1 anemia, and the other grade 3 anemia and grade 1 thrombocytopenia. This being said is relevant to note that two other dogs presented middle to severe cytopenias (one with grade 1 anemia and grade 4 thrombocytopenia and the other with grade 2 anemia and grade 2 thrombocytopenia) in the absence of bone marrow infiltration. This reinforces the principle previously stated in different articles, which report that peripheral blood and bone marrow findings are not always compatible (29, 30). The cytopenias encountered in dogs without bone marrow involvement were suspected to be of immune origin, which is a common cause of cytopenias found in dogs with lymphoma (29-31). Owing to this poor association between bone marrow affliction and hematological abnormalities, CBC results should not be used as an independent screening test for the presence of lymphoma in the bone marrow for staging canine lymphoma (29, 30).
Conclusion
In summary, it becomes clear that the discrepancies observed in stage allocation using different staging methods emphasize the need for a cautious comparison among published studies without consistent application of standardized staging tests. To ensure the robustness of research findings and comparison among studies, there is a need for the establishment of a unified and widely accepted staging protocol for canine lymphoma. This development will empower future research to meticulously scrutinize the impact of the clinical stage on both treatment modalities and the prognosis of canine lymphoma.
Acknowledgements
This work was financed by National Funds (FCT/MCTES, Fundação para a Ciência e a Tecnologia and Ministério da Ciência, Tecnologia e Ensino Superior) under the project UIDB/00772/2020. The authors also want to acknowledge the support received by projects UIDB/00211/2020 and LA/P/0059/2020, from FCT/MCTES.
Footnotes
Authors’ Contributions
FLQ and MTP conceived and designed the study; JB and IP followed the clinical cases and performed data collection; MTP created the database; FLQ and MTP performed data analysis; MTP wrote the draft version of the manuscript; MTP, JB, IP, and FLQ reviewed and approved the submitted version.
Conflicts of Interest
None of the authors of this paper has a financial or personal conflict of interest to declare. The present study resulted from a Master Dissertation performed by MTP under supervision of FLQ and JB (http://hdl.handle.net/10348/8814).
- Received January 2, 2024.
- Revision received February 7, 2024.
- Accepted February 8, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.